To identify unknown loci determining alcoholism, we are testing for linkage or association between genetic markers and behavioral phenotypes. The probability of establishing linkage or association is being maximized by 1) focusing on human alcoholism with impulsivity/aggressivity as a prominent accompanying behavioral trait, 2) utilizing mouse genetic models and 3) using a larger panel of markers than has heretofore been available. Markers include DNA polymorphisms defined by probes specific for the Y chromosome. In addition, we have mapped large panels of new human and mouse protein polymorphisms by two-dimensional electrophoresis. We have identified in humans more than 40 independent polymorphic loci expressed among the abundant proteins of lymphocytes, fibroblasts, erythrocytes and serum. A third of these variant human loci have been chromosomally mapped. These loci are being used for linkage analysis in families with alcoholism. In the mouse, we have identified 14 brain polypeptide variants and have preliminarily correlated one of these, which we mapped to chromosome 1, with ethanol intake (preference). We have chromosomally mapped four variant mouse brain polypeptides and identified twelve of the proteins visualized by two-dimensional electrophoresis of mouse brain. For correlative behavioral studies in collaboration, we have typed thirty strains of inbred mice at the 14 variant loci. We are studying the known enzymes of ethanol metabolism, principally alcohol and aldehyde dehydrogenases (ADH and ALDH). In this work, we have resolved on both analytic and preparative scales three human class III ADHs with a new method: immobiline isoelectric focusing. At the DNA level, we have completed a linkage analysis between the three Class I ADH genes located on chromosome 4 as well as a preliminary survey of the ADH and ALDH genes present in the hominoid primates. We have shown that the only ADH present in brain is Class III ADH and have purified, characterized, and immunohistochemically mapped the human brain enzyme.